Hanseok Ko
Papers
14
Total Citations
83
H-Index
6
About
Hanseok Ko is a leading researcher in robotics and human-robot interaction, with a focus on enabling robots to perceive and respond to their environments through advanced acoustic and visual processing. His work spans key areas including acoustic event recognition, robust speech interfaces, and semantic segmentation for unstructured environments. Ko's major contributions include developing acoustic filterbanks and post-filters that allow cleaning robots to perform surveillance by recognizing events despite ego-noise, and integrating acoustic echo cancellation with adaptive beamforming for full-duplex speech interaction. His research on gradient-based local affine invariant features has advanced mobile robot localization in indoor settings, while his recent work on memory-based semantic segmentation addresses the challenging problem of off-road navigation, achieving 13 citations. With over 70 citations across his most-cited papers, Ko's innovations have practical impacts on cleaning robots, security systems, and autonomous vehicles. Notably, his 2015 paper on acoustic event filterbanks for cleaning robots (14 citations) demonstrates how everyday devices can be repurposed for security, reflecting his talent for solving real-world engineering problems through elegant signal processing and machine learning solutions.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9Robust sound source localization using a Wiener filter4 citations · 2013
- 10